Skip to main content

Adaptive ADAS to support incapacitated drivers Mitigate Effectively risks through tailor made HMI under automation

Periodic Reporting for period 3 - ADASANDME (Adaptive ADAS to support incapacitated drivers Mitigate Effectively risks through tailor made HMI under automation)

Reporting period: 2019-09-01 to 2020-02-29

The ADAS&ME project (“Adaptive ADAS to support incapacitated drivers & Mitigate Effectively risks through tailor made HMI under automation”) will develop Advanced Driver Assistance Systems that adapts to the driver’s/rider’s state and the situational/environmental context when control is transferred between the vehicle and the driver/rider, ensuring a safer and more efficient road usage. The project in centred on seven use cases involving cars, trucks, buses and motorcycles. Each use case defines several societal benefits. For example, the combination of driver state monitoring and automated functions can effectively manage high risk situations, thereby reducing the number of traffic crashes due to driver error. Functions addressing other specific use cases such as long-haul trucking or bus driving will have a significant impact to emission reduction and more efficient transport. Experimental research will be carried out to develop and evaluate driver state monitoring algorithms, HMI designs and automation transitions. Robust algorithms for detection/prediction of fatigue/sleepiness, stress, inattention, emotions, thermal fatigue, faint and rest will be developed, making use of existing and novel sensing technologies, taking into account traffic and weather conditions via V2X as well as the individual driver’s physiology and driving behaviour. The detection/prediction results, along with the severity of scenarios, are used to design multimodal and adaptive warning and intervention strategies. The final outcome is the successful fusion of the developed elements into an integrated driver/rider state monitoring system, able to both be utilized in and be supported by vehicle automation of Levels 1 to 4. The system will be validated with a wide pool of drivers/riders under simulated and real road conditions and under different driver/rider states; with the use of 2 cars (1 conventional, 1 electric), 1 truck, 2 PTWs and 1 bus demonstrators. The expected impacts of ADAS&ME on mobility, congestion and safety are significant.
Identification of relevant user groups and stakeholders took place at start and a combination of different approaches was used to gather those needs and wishes by using questionnaires, focus groups and workshops. A SoA and benchmarking took place and a theoretical approach for the driver/rider state modelling was developed. Finally, a multi criteria analysis took place to select the final use cases and the prioritized scenarios that sets the framework for the development and evaluations. In total 7 use cases were selected including truck, car, motorbike and bus. The system architecture was developed. The work identified the main building blocks, the foreseen interactions and the communication between each component. Specifications for all included sensors were defined. The work on situation/environment awareness was decided to incorporate several information layers, taking the driver state, weather, traffic flow, surrounding traffic and HD Map geometry into account. Regarding the driver state algorithms, the algorithmic framework for has been set, definitions and golden standards have been identified and agreed upon. Existing data will be used for algorithm development, but for some states new data collections were needed. In total 6 data collections took place. The definition of the personalisation system was defined and specified. The HMI framework in general and per use case have been defined, including UML diagrams and HMI elements for all use cases. An overview about automated functions was done. In addition, a cloud-based platform for the system integration framework was developed. This will ensure the availability of the HMI, sensors, algorithms and environmental monitoring functions when integrating the full system in the various simulators and vehicle demonstrators. The framework of the evaluations was developed and a review of existing impact assessment methodologies of ADAS systems has been realised with a result of six impact areas important for economic and social impact assessment. Based on this the theoretical framework for each impact area have been identified, but also a synthesis of the legislative impact and a match with what is needed for the impact assessment. In addition, a pre-market analysis has been done.
As an overall vision ADAS&ME targets to develop Advanced Driver Assistance Systems (ADAS) that incorporate driver/rider state, situational/ environmental context and adaptive interaction, to automatically transfer control between vehicle and driver/rider and thus ensure safer and more efficient road usage for all vehicle types (conventional and electric car, truck, bus, motorcycle). ADAS&ME targets the detection of driver state through the use of unobtrusive sensors, installed in the seat, steering wheel or at the dashboard for the car/track use cases. For the motorcycle use cases the rider state is achieved through integrated sensors at the Personal Protection Equipment (PPE) of the rider. Driver/rider state monitoring will go beyond state of the art when it comes to multi-sensor integrated monitoring and fusion of different driver states, as well as to the development of new algorithms that combine driver state, vehicle state, environmental context and driver/rider background info. This is important to develop a system with an optimised HMI that is both reliable and useful by the drivers. ADAS&ME will use a generic user-oriented methodology where the decision about what to do to prevent or mitigate critical situations will be adapted to the current situation and the environmental context. ADAS&ME focuses on handovers and takeovers in safety relevant situations. New here is that the HMI logic considers a change of driver/rider state within an automation level as well as in the transitions between automation levels. ADAS&ME will use new technologies and outcomes from existing projects and merge them with the driver state information to develop new innovative HMI strategies.Those solutions are a step towards tailor-made information which the driver is capable of processing in continuously changing conditions on different TRL depending on the UC (for example motorcycle automation is still at its infancy therefore start TRL is 2). ADAS&ME will consider the comprehensive integration of driver/rider state with automated functions, and hence will optimise the overall combination of driver/rider and vehicle to drive towards the goal of safer driving, with consequent reduction of fatalities, injuries and their huge social costs (human and administrative).
Autoliv steering wheel
Scania truck
Busdriver during data collection
Use Cases
Example of data collection - Scania